Figure 3

Variability Generation Work Flow (VGWF). Starting from the multialignment of all the genomes stored in the database and of continent-specific subsets performed with MAFFT program, three variability analyses are executed separately: nucleotide variability values for both the entire set of genomes and the continent-specific subsets are estimated through SiteVar program applied on real dataset (right hand side of the Work Flow) and on 100 simulated multialignments obtained through an automatic procedure for each continent-specific genomes dataset (in the middle of the Work Flow); aminoacid variability data (left hand side of the Work Flow) are produced by applying MitVarProt software to the 13 multialigned mitochondrial protein coding genes, automatically selected from the entire starting nucleotide multialignment and translated into aminoacid sequences through TRANSEQ program; the data produced through this three procedures are inserted in NT_VARIABILITY, NT_VARIABILITY_SIM and AA_VARIABILITY tables of the relational HmtDB structure respectively.